0095-1137/78/0007-0052$02.00/0 Vol. 7, No. 1
JOURNAL OF CLINICAL MICROBIOLOGY, Jan. 1978, p. 52-54
Printed in U.S.A.
Copyright ( 1978 American Society for Microbiology
New Centrifugation Blood Culture Device G. L. DORN* AND K. SMITH Department of Microbiology, Leland Fikes Research Institute, and Granville C. Morton Cancer and Research Hospital, Divisions of Wadley Institutes of Molecular Medicine, Dallas, Texas 75235
Received for publication 13 July 1977
A single-tube blood culture device designed for centrifugation in a tabletop centrifuge is described. Reconstruction experiments using 21 different organisms and human donor blood indicate that excellent recovery can be obtained by centrifugation for 30 min at 3,000 x g. of the centrifuge rotor used. When this type of stopper is used, the high-density Fluorinert covers the entire stopper surface during centrifugation and thereby keeps the organism from becoming entrapped in the spaces between the glass wall and the stopper. The stopper was prepared by adding 1.1 ml of Sylgard 198 (Dow Coming Corp.) on top of a rubber stopper, and the tubes were held at an angle of 340 for 24 h at room temperature. After the angle stoppers were formed, the two solutions were added. The tubes were then evacuated to 35 mm of Hg, and the top stoppers were inserted. Sterility controls were done on a random selection of 5% of the tubes from each lot. Of the 4,000 tubes assembled to date, no contaminated tubes have been found. To determine the optimal centrifugation force and time, 21 different organisms were seeded into 21-dayold human donor citrate-phosphate-dextrose-blood (Wadley Blood Bank, Dallas, Tex.). The rubber stopper surfaces of the tube were scrubbed with 2% aqueous iodine and 70% isopropanol before insertion of a sterile needle. Seven milliliters of seeded blood, containing between 100 and 2,000 colony-forming units, was aseptically added into the tube. The contents were vigorously mixed, and the tube was placed in the centrifuge, angle stopper down with the angular plane facing toward the center of the rotor. After centrifugation at the desired gravity and time, the vessel was removed, and a sterile cotton-plugged needle was inserted into the top of the stopper to serve as a vent. A 10-ml syringe was fitted with a 21gauge, 1.5-inch (ca. 3.8-cm) needle. This needle was inserted through the bottom of the angle stopper to its upper limit, and the majority of the supernatant (approximately 7 ml) was slowly removed. The remaining 1.5 ml of solution was vigorously mixed and removed with a 2-ml syringe. The contents within this syringe were equally dispensed onto five plates containing suitable growth medium and incubated under the appropriate atmospheres (Tables 1 and 2). A 1ml sample of the supernatant was also plated to determine the total count remaining in the supernatant. Control counts were made of each organism by preparing dilutions of the inoculum into an appropriate solution (Tables 1 and 2). From this data it was possible to obtain a mean (x) and standard deviation (SD) of the relative recovery in the concentrate as compared with the total count in the centrifugation
Previous studies suggested that centrifugation can be used to provide the clinical laboratory with a rapid, quantitative, and versatile blood culture technique (1, 2). The original two-tube method was based upon the principle of centrifugation of microorganisms on or into a stabilizing density layer after effective lyses of the erythrocytes (2). This method showed considerable promise in that substantial increases in the number of positive cultures were obtained (13 versus 7%), and the time required to obtain a pure isolate was considerably shorter. Land et al. were able to demonstrate that the centrifuge technique was especially valuable in the detection of disseminated fungal infections (3). Nevertheless, this technique was cumbersome and yielded an undesirably high contamination rate (9.3 versus 1.8% for the bottle technique). In addition, excessively high centrifugation forces (6,000 x g) in a swinging-bucket rotor were required to achieve acceptable recovery for several microorganisms. In an effort to resolve these difficulties, a new vehicle and process has been developed that more readily lends itself to the clinical laboratory. This report describes the new methodology and provides data on the recovery of 21 different organisms from blood as a function of relative centrifugal force and time in an angle-rotor tabletop centrifuge (Sorval GLC-2, SP-X rotor; Ivan Sorvall, Inc., Norwalk, Conn.). MATERILS AND METHODS A schematic of the new centrifugation vehicle is presented in Fig. 1. It is essentially a double-stopper, evacuated Pyrex tube that contains 0.3 ml of a highdensity hydrophobic cushion (Fluorinert, 3-M Corp.) and 1.2 ml of an aqueous solution. This latter solution contains 0.08% Dow Corning Anti-Foam B, 10 mg of Solryth (Hoffman-La Roche, Inc., Nutley, N.J.), 5 mg of sodium polyanetholsulfonate (Hoffiman-La Roche), and 15 yg of sodium thioglycolate (Nutritional Biochemicals Corp., Cleveland, Ohio). The bottom stopper has an angular plane complementary to the angle
52
VOL. 17, 1978
53
CENTRIFUGATION BLOOD CULTURE DEVICE
vessel. In addition, a survival index (K value) was determined as follows: K = total count in supernatant + total count in concentrate/total count in the control (saline or broth). A K = 1 indicates that no microorganisms were lost through the killing action of the blood,
Vacuum ch"am
I/0
I //I .
-7I
HeNulyzie ad inticsagulatig agedt
Cowls ,...
FIG. 1. Cross section of the centrifugation vehicle.
the other components in the tube, or the manipulation process itself.
RESULTS
Tables 1 and 2 summarize the results obtained using three different combinations of gravity and time. The best overall average recovery (93%) was obtained at 3,000 x g for 30 min. With this combination only three organisms gave less than 80% recovery, namely, Pseudomonas aeruginosa (77%), Bacteroides fragilis (75%), and Neisseria meningitidis (59%). The other combinations gave lower overall recoveries (86 and 85%, respectively) and were generally less efficient with several of the other organisms tested, namely, Brucella suis, Bacteroides melaninogenicus, and Salmonella typhi. It should be emphasized that an average recQvery of 93% does not mean that the organism was not detected 7% of the time. On the contrary, the data indicate that if 100 bacteria were present in the blood sample, on the average, 93 organisms would be recovered. With the exception of Staphylococcus aureus, the survival value (K) ranged from 0.8 to 1.1, indicating that no toxicity was observed with the present vehicle and proc-
TABLE 1. Percent recovery of various microorganisms from blood by centrifugation Recoverya
1,500 RCF
3,000 RCF' (min) Organism
Avg
CFUb/
__
_-
30
15
30
x + SD(%)
i ±SD(%)
i ±SD (%)
97± 2
99± 1
98± 1
tube
Aspergillus nidulans Candida albicans Cryptococcus neoformans Brucella suis (ATCC 4312) Citrobacter freundii Enterobacter cloacae Escherichia coli Haemophilus influenzae (ATCC 19418)
Klebsiellapneumoniae Proteus vulgaris (ATCC 6380) Pseudomonas aeruginosa
Salmonella typhi (ATCC 6539) Neisseria meningitidis (ATCC 13077) Staphylococcus aureus Streptococcus pneumoniae (ATCC 6301) Streptococcus pyogenes (ATCC 19616) Bacteroides fragilis (ATCC 23745) Bacteroides melaninogenicus (ATCC 25845) Clostridium sporogenes (ATCC 500310) Fusobacterium nucleatum (ATCC 25534) Listeria monocytogenes (ATCC 984) Avg recovery a Percent recovery given as i ± SD. Six tubes were and organism. b CFU, Colony-forming unit. c RCF, Relative centrifugal force.
1,433 1,085 669 600 700 250 742 300
200 100
950 600
1,568
982 500
100
99± 2
99± 1
100 82 ± 2 95 ± 2 94 ± 5 95± 2
99± 1 68 ± 4 97± 2 72 ± 7 81 ± 3
100 72 ± 5 75 ± 10 92 ± 4 88± 3
98 ± 3 98 ± 4 95 ± 5 77 ± 8 96 ± 5 59 ± 13
86± 5
82 ± 6
98± 3
98± 2
94 ± 92 ± 78 ± 45 ±
11
14 19 4 100± 1
91 ± 7
68 ± 7 77 ± 14 47 ± 3 99± 1 92 ± 5 99± 2
300
100 90 ± 6 100± 1
1,351
75 ± 2
93 ± 10 63 ± 4
1,980
86 + 2
750
98 ± 2 98 1
90± 9
57 ± 4 65 ± 3 94 ± 3
2,000
100± 1 97 6
99± 1 90± 6
99± 1 90± 5
86 ± 6
85 ± 4
850 -
processed
93
for each
4
unique
87 ± 7
combination of RCF, time,
54
J. CLIN. MICROBIOL.
DORN AND SMITH
TABLE 2. Conditions for recovery of microorganisms from blood by centrifugation
Organism AspergiUus nidulans
Dilutant
Growth medium Sabouraud agar (Difco)
Normal saline
Plate me-
conditions
Sabouraud
360C, aerobic,
agar
Candida albicans Cryptococcus neoformans Citrobacter freundii Enterobacter cloacae Escherichia coli Klebsiella pneumoniae Proteus vulgaris Pseudomonas aeruginosa Salmonella typhi Staphylococcus aureus
1% Dextrose, 1% peptone, and nutrient broth (Difco)
Brain heart infusion broth
Brucella suis Neisseria meningitidis Listeria monocytogenes
Brain heart infusion
Brain heart infusion
Streptococcus pneumoniae Streptococcus pyogenes
Todd-Hewitt (Difco)
Bacteroides fragilis Bacteroides melaninogenicus Clostridium sporogenes Fusobacterium nuckatum
Chopped meat (Difco)
Haemophilus influenzae
Brain heart infusion and supplement B (Difco)
(Difco)
Todd-Hewitt
Atmospheric
dium
Blood agar
(BBL)
24 h
360C, aerobic, 24 h
Chocolate agar
36°C, 5% C02,
Blood agar
36'C, 5% C02,
48 h
48 h Prereduced brain heart infusion
Chocolate agar
360C, BBL anaerobic jar, 48 h
ess for the organisms tested. One possible explanation for the 1.4 K value for S. aureus is that the inoculum contained aggregates of several colony-forming units that were dispersed during the centrifugation procedure. No contamination by a second organism was observed during this study. However, these experiments were done under carefully controlled research conditions.
DISCUSSION The single-tube vehicle described here required fewer entries and less manipulation than its two-tube predecessor. Furthermore, it is adaptable to a nonrefrigerated tabletop centrifuge (Sorval GLC or Beckman TJ-6) fitted with a multiple-sample angle rotor. The reconstruction data indicated that, at 3,000 x g for 30 min, the efficiency of the new vehicle is far superior to the earlier two-tube system that required a
Brain heart infusion and supplement B
Chocolate agar
360C, 5% C02, 48 h
swinging-bucket rotor (2). A clinical trial is currently in progress to ascertain the relative efficiency and contamination rate of this system in a clinical environment. ACKNOWLEDGMENTS This research was supported by funds from the Wadley Guild, The Leukemia Association of North Central Texas, and Medical Research Inc., of Dallas, Tei. LITERATURE CITED 1. Dorn, G. L., G. G. Burson, and J. R. Haynes. 1976. Blood culture technique based on centrifugation: clinical evaluation. J. Clin. Microbiol. 3:258-263. 2. Dorn, G. L, J. R. Haynes, and G. G. Burson. 1976. Blood culture technique based on centrifugation: developmental phase. J. Clin. Microbiol. 3:251-257. 3. Land, G. A., G. L Dorn, and J. M. Hill. 1977. The isolation of disseminated Coccidioides immitis by an improved blood culture technique, p. 19-30. In Coccidioidomycosis: current clinical and diagnostic status.
JOURNAL OF CLINICAL MICROBIOLOGY, Jan. 1978, p. 52-54
Printed in U.S.A.
Copyright ( 1978 American Society for Microbiology
New Centrifugation Blood Culture Device G. L. DORN* AND K. SMITH Department of Microbiology, Leland Fikes Research Institute, and Granville C. Morton Cancer and Research Hospital, Divisions of Wadley Institutes of Molecular Medicine, Dallas, Texas 75235
Received for publication 13 July 1977
A single-tube blood culture device designed for centrifugation in a tabletop centrifuge is described. Reconstruction experiments using 21 different organisms and human donor blood indicate that excellent recovery can be obtained by centrifugation for 30 min at 3,000 x g. of the centrifuge rotor used. When this type of stopper is used, the high-density Fluorinert covers the entire stopper surface during centrifugation and thereby keeps the organism from becoming entrapped in the spaces between the glass wall and the stopper. The stopper was prepared by adding 1.1 ml of Sylgard 198 (Dow Coming Corp.) on top of a rubber stopper, and the tubes were held at an angle of 340 for 24 h at room temperature. After the angle stoppers were formed, the two solutions were added. The tubes were then evacuated to 35 mm of Hg, and the top stoppers were inserted. Sterility controls were done on a random selection of 5% of the tubes from each lot. Of the 4,000 tubes assembled to date, no contaminated tubes have been found. To determine the optimal centrifugation force and time, 21 different organisms were seeded into 21-dayold human donor citrate-phosphate-dextrose-blood (Wadley Blood Bank, Dallas, Tex.). The rubber stopper surfaces of the tube were scrubbed with 2% aqueous iodine and 70% isopropanol before insertion of a sterile needle. Seven milliliters of seeded blood, containing between 100 and 2,000 colony-forming units, was aseptically added into the tube. The contents were vigorously mixed, and the tube was placed in the centrifuge, angle stopper down with the angular plane facing toward the center of the rotor. After centrifugation at the desired gravity and time, the vessel was removed, and a sterile cotton-plugged needle was inserted into the top of the stopper to serve as a vent. A 10-ml syringe was fitted with a 21gauge, 1.5-inch (ca. 3.8-cm) needle. This needle was inserted through the bottom of the angle stopper to its upper limit, and the majority of the supernatant (approximately 7 ml) was slowly removed. The remaining 1.5 ml of solution was vigorously mixed and removed with a 2-ml syringe. The contents within this syringe were equally dispensed onto five plates containing suitable growth medium and incubated under the appropriate atmospheres (Tables 1 and 2). A 1ml sample of the supernatant was also plated to determine the total count remaining in the supernatant. Control counts were made of each organism by preparing dilutions of the inoculum into an appropriate solution (Tables 1 and 2). From this data it was possible to obtain a mean (x) and standard deviation (SD) of the relative recovery in the concentrate as compared with the total count in the centrifugation
Previous studies suggested that centrifugation can be used to provide the clinical laboratory with a rapid, quantitative, and versatile blood culture technique (1, 2). The original two-tube method was based upon the principle of centrifugation of microorganisms on or into a stabilizing density layer after effective lyses of the erythrocytes (2). This method showed considerable promise in that substantial increases in the number of positive cultures were obtained (13 versus 7%), and the time required to obtain a pure isolate was considerably shorter. Land et al. were able to demonstrate that the centrifuge technique was especially valuable in the detection of disseminated fungal infections (3). Nevertheless, this technique was cumbersome and yielded an undesirably high contamination rate (9.3 versus 1.8% for the bottle technique). In addition, excessively high centrifugation forces (6,000 x g) in a swinging-bucket rotor were required to achieve acceptable recovery for several microorganisms. In an effort to resolve these difficulties, a new vehicle and process has been developed that more readily lends itself to the clinical laboratory. This report describes the new methodology and provides data on the recovery of 21 different organisms from blood as a function of relative centrifugal force and time in an angle-rotor tabletop centrifuge (Sorval GLC-2, SP-X rotor; Ivan Sorvall, Inc., Norwalk, Conn.). MATERILS AND METHODS A schematic of the new centrifugation vehicle is presented in Fig. 1. It is essentially a double-stopper, evacuated Pyrex tube that contains 0.3 ml of a highdensity hydrophobic cushion (Fluorinert, 3-M Corp.) and 1.2 ml of an aqueous solution. This latter solution contains 0.08% Dow Corning Anti-Foam B, 10 mg of Solryth (Hoffman-La Roche, Inc., Nutley, N.J.), 5 mg of sodium polyanetholsulfonate (Hoffiman-La Roche), and 15 yg of sodium thioglycolate (Nutritional Biochemicals Corp., Cleveland, Ohio). The bottom stopper has an angular plane complementary to the angle
52
VOL. 17, 1978
53
CENTRIFUGATION BLOOD CULTURE DEVICE
vessel. In addition, a survival index (K value) was determined as follows: K = total count in supernatant + total count in concentrate/total count in the control (saline or broth). A K = 1 indicates that no microorganisms were lost through the killing action of the blood,
Vacuum ch"am
I/0
I //I .
-7I
HeNulyzie ad inticsagulatig agedt
Cowls ,...
FIG. 1. Cross section of the centrifugation vehicle.
the other components in the tube, or the manipulation process itself.
RESULTS
Tables 1 and 2 summarize the results obtained using three different combinations of gravity and time. The best overall average recovery (93%) was obtained at 3,000 x g for 30 min. With this combination only three organisms gave less than 80% recovery, namely, Pseudomonas aeruginosa (77%), Bacteroides fragilis (75%), and Neisseria meningitidis (59%). The other combinations gave lower overall recoveries (86 and 85%, respectively) and were generally less efficient with several of the other organisms tested, namely, Brucella suis, Bacteroides melaninogenicus, and Salmonella typhi. It should be emphasized that an average recQvery of 93% does not mean that the organism was not detected 7% of the time. On the contrary, the data indicate that if 100 bacteria were present in the blood sample, on the average, 93 organisms would be recovered. With the exception of Staphylococcus aureus, the survival value (K) ranged from 0.8 to 1.1, indicating that no toxicity was observed with the present vehicle and proc-
TABLE 1. Percent recovery of various microorganisms from blood by centrifugation Recoverya
1,500 RCF
3,000 RCF' (min) Organism
Avg
CFUb/
__
_-
30
15
30
x + SD(%)
i ±SD(%)
i ±SD (%)
97± 2
99± 1
98± 1
tube
Aspergillus nidulans Candida albicans Cryptococcus neoformans Brucella suis (ATCC 4312) Citrobacter freundii Enterobacter cloacae Escherichia coli Haemophilus influenzae (ATCC 19418)
Klebsiellapneumoniae Proteus vulgaris (ATCC 6380) Pseudomonas aeruginosa
Salmonella typhi (ATCC 6539) Neisseria meningitidis (ATCC 13077) Staphylococcus aureus Streptococcus pneumoniae (ATCC 6301) Streptococcus pyogenes (ATCC 19616) Bacteroides fragilis (ATCC 23745) Bacteroides melaninogenicus (ATCC 25845) Clostridium sporogenes (ATCC 500310) Fusobacterium nucleatum (ATCC 25534) Listeria monocytogenes (ATCC 984) Avg recovery a Percent recovery given as i ± SD. Six tubes were and organism. b CFU, Colony-forming unit. c RCF, Relative centrifugal force.
1,433 1,085 669 600 700 250 742 300
200 100
950 600
1,568
982 500
100
99± 2
99± 1
100 82 ± 2 95 ± 2 94 ± 5 95± 2
99± 1 68 ± 4 97± 2 72 ± 7 81 ± 3
100 72 ± 5 75 ± 10 92 ± 4 88± 3
98 ± 3 98 ± 4 95 ± 5 77 ± 8 96 ± 5 59 ± 13
86± 5
82 ± 6
98± 3
98± 2
94 ± 92 ± 78 ± 45 ±
11
14 19 4 100± 1
91 ± 7
68 ± 7 77 ± 14 47 ± 3 99± 1 92 ± 5 99± 2
300
100 90 ± 6 100± 1
1,351
75 ± 2
93 ± 10 63 ± 4
1,980
86 + 2
750
98 ± 2 98 1
90± 9
57 ± 4 65 ± 3 94 ± 3
2,000
100± 1 97 6
99± 1 90± 6
99± 1 90± 5
86 ± 6
85 ± 4
850 -
processed
93
for each
4
unique
87 ± 7
combination of RCF, time,
54
J. CLIN. MICROBIOL.
DORN AND SMITH
TABLE 2. Conditions for recovery of microorganisms from blood by centrifugation
Organism AspergiUus nidulans
Dilutant
Growth medium Sabouraud agar (Difco)
Normal saline
Plate me-
conditions
Sabouraud
360C, aerobic,
agar
Candida albicans Cryptococcus neoformans Citrobacter freundii Enterobacter cloacae Escherichia coli Klebsiella pneumoniae Proteus vulgaris Pseudomonas aeruginosa Salmonella typhi Staphylococcus aureus
1% Dextrose, 1% peptone, and nutrient broth (Difco)
Brain heart infusion broth
Brucella suis Neisseria meningitidis Listeria monocytogenes
Brain heart infusion
Brain heart infusion
Streptococcus pneumoniae Streptococcus pyogenes
Todd-Hewitt (Difco)
Bacteroides fragilis Bacteroides melaninogenicus Clostridium sporogenes Fusobacterium nuckatum
Chopped meat (Difco)
Haemophilus influenzae
Brain heart infusion and supplement B (Difco)
(Difco)
Todd-Hewitt
Atmospheric
dium
Blood agar
(BBL)
24 h
360C, aerobic, 24 h
Chocolate agar
36°C, 5% C02,
Blood agar
36'C, 5% C02,
48 h
48 h Prereduced brain heart infusion
Chocolate agar
360C, BBL anaerobic jar, 48 h
ess for the organisms tested. One possible explanation for the 1.4 K value for S. aureus is that the inoculum contained aggregates of several colony-forming units that were dispersed during the centrifugation procedure. No contamination by a second organism was observed during this study. However, these experiments were done under carefully controlled research conditions.
DISCUSSION The single-tube vehicle described here required fewer entries and less manipulation than its two-tube predecessor. Furthermore, it is adaptable to a nonrefrigerated tabletop centrifuge (Sorval GLC or Beckman TJ-6) fitted with a multiple-sample angle rotor. The reconstruction data indicated that, at 3,000 x g for 30 min, the efficiency of the new vehicle is far superior to the earlier two-tube system that required a
Brain heart infusion and supplement B
Chocolate agar
360C, 5% C02, 48 h
swinging-bucket rotor (2). A clinical trial is currently in progress to ascertain the relative efficiency and contamination rate of this system in a clinical environment. ACKNOWLEDGMENTS This research was supported by funds from the Wadley Guild, The Leukemia Association of North Central Texas, and Medical Research Inc., of Dallas, Tei. LITERATURE CITED 1. Dorn, G. L., G. G. Burson, and J. R. Haynes. 1976. Blood culture technique based on centrifugation: clinical evaluation. J. Clin. Microbiol. 3:258-263. 2. Dorn, G. L, J. R. Haynes, and G. G. Burson. 1976. Blood culture technique based on centrifugation: developmental phase. J. Clin. Microbiol. 3:251-257. 3. Land, G. A., G. L Dorn, and J. M. Hill. 1977. The isolation of disseminated Coccidioides immitis by an improved blood culture technique, p. 19-30. In Coccidioidomycosis: current clinical and diagnostic status.
